Chemical Stability of Anion Exchange Membranes for Alkaline Fuel Cells

Chapter 14. Chemical Stability of Anion ... cell (1). The AFC technology was first patented by Reid in 1903 (2), demonstrated by Bacon in 1932 (3), ...
0 downloads 0 Views 1004KB Size
Chapter 14

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Chemical Stability of Anion Exchange Membranes for Alkaline Fuel Cells Yuesheng Ye and Yossef A. Elabd* Department of Chemical and Biological Engineering, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104 *[email protected]

Solid-state alkaline fuel cells (AFCs), which utilize anion exchange membranes (AEMs) as their electrolytes, have the potential to provide society with low-cost, long-lasting renewable portable power. However, the chemical stability of AEMs poses a critical challenge that limits the wide scale use of AFCs. This chapter reviews literature findings on the chemical stability of recently developed hydroxide conducting AEMs with various backbone and cation chemical structures, where covalently attached cations in AEMs, include ammonium, phosphonium, sulfonium, guanidinium, imidazolium, pyridinium, quaternized 1,4-diazabicyclo(2,2,2)octane (DABCO), and piperidinium. However, limited chemical analysis has been conducted regarding the chemical stability of AEMs, therefore, this chapter will also discuss earlier work on the chemical stability of small molecule cation analogs in alkaline conditions. In the future, a more fundamental understanding of the chemical stability of AEMs will be required to adequately design robust solid-state AFCs.

Introduction The alkaline fuel cell (AFC), classified by electrolyte type (i.e., electrolytes that conduct hydroxide anions (OH-)), is among the oldest and most powerful fuel cell (1). The AFC technology was first patented by Reid in 1903 (2), demonstrated by Bacon in 1932 (3), utilized in NASA’s Apollo space missions in the mid-1960s (1), and developed by Kordesch for AFC-powered motorcycles (4). However, © 2012 American Chemical Society In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

the use of liquid electrolytes (e.g., KOH(aq)) in AFCs has significantly limited its wide scale use due to the durability issues encountered with electrolyte leakage and the precipitation of carbonate crystals, such as potassium carbonate (K2CO3), due to carbonation (i.e., exposure to CO2 in the fuel) (5). The acid fuel cell, however, mitigated problems associated with liquid electrolytes in the 1950s with the incorporation of solid-state electrolytes that can readily transport protons. This work led to the development of the proton exchange membrane fuel cell (PEMFC) and was utilized in NASA’s Gemini space missions (6). Furthermore, the development of Nafion®, a PEM with excellent properties developed by DuPont in 1960s (7), significantly accelerated research and development of PEMFCs for wide scale application in portable power (e.g., automobiles). Similarly, a robust solid-state membrane that can conduct hydroxide ions (i.e., anion exchange membrane (AEM)) could replace liquid electrolytes in AFCs. This would eliminate leakage and carbonate precipitation issues associated with liquid electrolytes and accelerate the development of long-lasting AFCs. To date, recent advances in the PEM counterpart, the alkaline AEM (8, 9), have significantly renewed interest in the AFC. The AEM AFC not only circumvents problems encountered with liquid-based alkaline electrolytes, but also provides significant advantages over PEMFCs, including enhanced electrokinetics that allows for the use of non-noble metal catalysts, reduced fuel crossover, and improved water management (10). Most importantly, a solid-state AFC using AEMs holds the promise of a low-cost, long-lasting fuel cell. Recently, a number of AEMs have been developed for the AFC (several examples shown in Table 1). However, the chemical stability of AEMs is still a critical and challenging factor that limits the wide scale use of solid-state AFCs. Specifically, the high nucleophilicity and basicity of OH- ions can trigger the degradation of a covalently attached counter cation, as well as the polymer backbone. Therefore, it is essentially important to understand the chemical stability of AEMs under various conditions. However, studies on the chemical stability of AEMs are still relatively scarce. Particularly, recent stability studies on AEMs for AFCs have mainly focused on evaluating ion exchange capacity (IEC) or ionic conductivity over a designated period of time after an AEM sample is immersed in a concentrated alkaline solution at a certain temperature (see literature results in Table 2). These results suggest that a change or lack thereof in IEC or ionic conductivity provides an understanding of the chemical stability of the AEM under a chosen test condition. However, this approach has several shortcomings. For instance, IEC or conductivity may not truly represent the chemical stability of the AEM in OH- form since hydroxide ions may quickly convert to carbonate (CO32-) and/or bicarbonate (HCO3-) ions in the presence of ambient air (i.e., carbonation due to ~ 400 ppm of CO2 in air) (11). Furthermore, this approach does not provide a quantitative chemical analysis or a deep understanding of chemical stability of AEMs, e.g., degradation mechanisms and kinetics.

234 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Table 1. Examples of AEMs for the AFC

235 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Table 2. Summary of chemical stability studies on AEMs for the AFC

Continued on next page.

236 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Table 2. (Continued). Summary of chemical stability studies on AEMs for the AFC

a ΔIEC and Δσ denote a negative change in ion exchange capacity and ionic conductivity, respectively. ΔNMR denotes the degree of degradation estimated from NMR spectroscopy results.

In contrast to the limited studies on chemical stability of AEMs, the chemical stability of small molecule organic cations in the presence of alkaline solutions have been extensively investigated in the past. Although the current status of AEM development for AFCs has been reviewed (1, 10, 12–15), including two recent comprehensive review articles (1, 15), the chemical stability of AEMs 237 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

was not thoroughly discussed in these review articles. Therefore, in this chapter, results on the chemical stability of AEMs, as well as the chemical stability of small molecule organic cations will be reviewed. Both non-cyclic cations (e.g., ammonium, phosphonium, sulfonium, guanidinium) and cyclic cations (e.g., imidazolium, pyridinium, quaternized 1,4-diazabicyclo(2,2,2)octane (DABCO), and piperidinium) will be discussed. A brief summary of the chemical stability of the polymer backbone in AEMs will also be discussed.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Chemical Stability of Ammonium-Based AEMs The chemical stability of ammonium-based AEMs has its origins in early work on using AEMs for desalination (37) and as ion exchangers (38, 39). Two major degradation reactions are generally accepted for ammonium-based AEMs: 1) a nucleophilic displacement (substitution) from the attack of OH- on the α carbons via a SN2 reaction resulting in two byproducts (amine and alcohol) (Scheme 1), and 2) an E2 (Hofmann) elimination from the abstraction of β hydrogen by OH- resulting in two byproducts (amine and alkene) (Scheme 2). Note that the SN2 reaction and E2 elimination occur in competition with one another, often leading to a mixture of byproducts. Different chemical structures associated with the ammonium cation may favor one degradation mechanism over the other. For instance, the steric hindrance of bulky chemical groups near the α carbon of the ammonium cation preferentially leads to the E2 elimination reaction. Alternatively, if the β hydrogen is not present, the degradation prefers a SN2 pathway. For example, the benzyltrimethylammonium cation with no β hydrogen undergoes a SN2 reaction and yields trimethylamine (65%) and benzyldimethylamine (35%) (see Scheme 1) (37).

Scheme 1. Nucleophilic displacement (substitution) of benzyltrimethylammonium cation (37)

238 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Scheme 2. E2 (Hofmann) elimination reaction of a quaternary ammonium cation (37) The development of ammonium-based AEMs have received the most attention for solid-state AFCs, as well as the study of chemical stability. Most ammonium-based AEMs (20–27) (See Tables 1 and 2) contain an arylammonium (e.g., benzylammonium) group. Although a few recent studies have quantified the degree of degradation using 1H NMR (27), most stability studies were based on the results of IEC or ionic conductivity (See Table 2). Table 2 shows test conditions of alkaline concentration, temperature, and immersion time in an alkaline solution ranging from 0 (water) to 10 M, 25 to 80 °C, and 2 to 2856 h, respectively. Early stability studies were conducted in water for long periods of time (e.g., 2856 h) (9), while more recent ones were conducted in concentrated alkaline solutions (21–28). Also, most stability tests were carried out at either a low temperature (< 60 °C) or a low alkaline concentration (~ 1 M), indicating that both high temperature (≥ 60 °C) and high alkaline concentration (≥ 1 M) will significantly accelerate degradation reactions. For example, at 6 M and 60°C, the benzyltrimethylammonium-based AEM experienced a 50% loss in conductivity after 2 h (30). Compared to the work on the chemical stability of ammonium-based AEMs over the past decade, the chemical stability of small molecule ammonium cations has been studied more extensively for more than a century. Work on the stability of the ammonium cation dates back to the 1850s when August Wilhelm von Hofmann pioneered the synthesis of ammonium and phosphonium bases (40, 41). Subsequently, the Hofmann (E2) elimination was named after the scientist for his work on amines and ammonium bases and organic phosphorus compounds. In addition to the E2 elimination, other major degradation mechanisms include the Stevens and Sommelet-Hauser rearrangement reactions. The Stevens rearrangement reaction (Scheme 3) was first discovered by Stevens and coworkers (42) in 1928 when treating phenacylbenzyldimethylammonium bromide with aqueous sodium hydroxide. The reaction was recognized as an intramolecular migration with a 1,2-rearrangment based on crossover experiments (43), and confirmed by 13C labeling (44). As for the reaction mechanism, it is clear that the rearrangement reaction begins with the abstraction of the α proton by the base to give an ylide intermediate and this was isolated by others (45). However, regarding the further reaction of the ylide intermediate, several mechanisms including ion-pair (44), concerted shift (46) and radical-pair (47) pathways were proposed. Note that the Stevens rearrangement of benzyltriammonium cation results in tertiary 239 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

benzylamines (Scheme 4) that are different from the products obtained from the nucleophilic displacement (SN2) pathway (Scheme 1).

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Scheme 3. Stevens rearrangement of an ammonium cation (42)

Scheme 4. Stevens rearrangement of benzyltrimethylammonium cation (42) Another mechanism, the Sommelet-Hauser rearrangement (Scheme 5), was first observed by Sommelet in 1937 with observations on the rearrangement of benzhydryltrimethylammonium hydroxide to o-benzylbenzyldimethylamine upon heating with concentrated sodium hydroxide (48). These results were confirmed with further investigations by Kantor and Hauser in 1951 (49). The Sommelet-Hauser rearrangement occurs only with certain quaternary benzyl ammonium cations. The reaction mechanism is simpler than the Stevens rearrangement, and was confirmed by intermediate isolation (50, 51) and 13C labeling experiments (52). For the benzyltrimethylammonium cation, both Stevens (Scheme 4) and Sommelet-Hauser (Scheme 5) rearrangement reactions are possible, and it was reported that the Stevens rearrangement is favored at high temperatures, while the Sommelet-Hauser is favored at lower temperatures (53).

Scheme 5. Sommelet–Hauser rearrangement of benzyltrimethylammonium cation (49) In addition to these rearrangement pathways, Pivovar and coworkers (54) recently reported that the direct degradation of an ylide intermediate results in alcohol and amine byproducts (Scheme 6), suggesting that the degradation pathway of nucleophilic displacement via a SN2 reaction (Scheme 1) is not the only pathway to produce alcohol and amine byproducts. 240 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Scheme 6. Ylide degradation pathway of benzyltrimethylammonium cation (54) Degradation studies on small molecule quaternary ammonium ions in alkaline conditions suggest that the degradation of ammonium-based AEMs is not a simple process and may include a number of degradation pathways, such as nucleophilic displacement (SN2), E2 (Hofmann) elimination, Stevens and Sommelet-Hauser rearrangements, and direct ylide degradation. However, the chemical stability of ammonium-based AEMs involving the Stevens and Sommelet-Hauser rearrangement reactions has not been experimentally investigated. Thus, more detailed and fundamental studies are necessary to gain better understanding of the possible degradation pathways and the extent of degradation for ammonium-based AEMs under various alkaline conditions.

Chemical Stability of Phosphonium-Based AEMs Compared to the work on ammonium-based AEMs, there are fewer investigations on phosphonium-based AEMs for the AFCs. This may be due to the much lower chemical stability of small molecule quaternary phosphonium cations compared to their ammonium analogs (55). For example, the instability of phosphonium-based AEMs was evidenced by a complete loss of water sorption capacity when poly(benzyltrialkylphosphonium chloride) ion exchange resins were exposed to 4% aqueous NaOH for 24 h (55). However, recent work showed that 2,4,6-trimethoxyphenyl bulky groups surrounding the phosphonium cation can significantly enhance chemical stability of phosphonium-based AEMs compared to quaternary ammonium-based AEMs (29). Ionic conductivity results on this AEM suggest high chemical stability after immersion in concentrated alkaline solutions (1-10 M) at room temperature for 48 h (see Table 2). The enhanced chemical stability was attributed to the strong electron-donating groups (o-methoxy (o-OCH3) and p-methoxy (p-OCH3)) in the benzene ring and the steric hindrance of the bulky trimethoxyphenyl groups in the cation (29). Although there is continued work on phosphonium-based AEMs (56, 57), their degradation mechanisms, in general, have not been well investigated. 241 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

In contrast, the degradation of small molecule organic phosphonium cations has been explored extensively by Hofmann and Cahours (58), Ingold and coworkers (59, 60), and other researchers, such as McEwen and coworkers (61, 62). The degradation of a quaternary phosphonium cation in alkaline conditions results in phosphine oxide and a hydrocarbon as byproducts and is referred to as the Cahours-Hofmann reaction (63). According to the work of McEwen and coworkers (61), the degradation of a phosphonium cation in alkaline conditions consists of the following steps (Scheme 7): 1) rapid, reversible attack by the hydroxide anion at a face of the phosphonium tetrahedron to produce a hydroxyphosphorane; 2) rapid, reversible removal of a proton from the hydroxyphosphorane to generate a phosphoranyl anion; 3) irreversible, rate-determining expulsion of a carbanion to form a tertiary phosphine oxide, where the expulsion and protonation of the carbanion probably occur simultaneously rather than as separated steps. The degradation of the benzyltrimethylphosphonium cation (Scheme 7) results in toluene and trimethylphosphine oxide as byproducts. Therefore, one would expect that the degradation of benzyltrimethylphosphonium-based AEMs would yield poly(4-methyl styrene).

Scheme 7. Degradation of benzyltrimethylphosphonium cation (61)

Note that the significant difference in chemical stability between the benzyltri(2,4,6-trimethoxyphenyl) phosphonium cation and a more conventional quaternary phosphonium cation suggests that the chemical groups attached to the phosphonium cation can greatly affect the chemical stability of the cation. For example, for the benzyltriarylphosphonium cation, electron-withdrawing groups, such as m-Cl, p-Cl and m-OCH3 (methoxy), can accelerate the degradation reaction. This acceleration was attributed to an increase in the concentration of the intermediate (Scheme 7) (64). In contrast, electron-withdrawing groups, such as p-OCH3 and o-OCH3, can inhibit the degradation reaction and enhance the chemical stability. For example, the introduction of a single o-OCH3 group into the benzyltriphenylphosphonium structure reduces the rate of alkaline cleavage to only 1/37 of its value compared to the unsubstituted benzyltriphenylphosphonium cation. The introduction of one more o-OCH3 group into the same benzene ring (i.e., benzyl(2,6-dimethoxyphenyl)diphenylphosphonium cation) can 242 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

reduce the degradation rate to 1/380 compared to the unsubstituted cation. Notice that the newly developed phosphonium-based AEMs, containing three 2,4,6-trimethoxyphenyl groups with electron-donating groups of o-OCH3 and p-OCH3, greatly enhance the stability of the phosphonium cation (29). In other words, the chemical stability can be significantly altered by designing the chemistry around the cation to inhibit the degradation reaction.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Chemical Stability of Sulfonium-Based AEMs There are only a few reports on the chemical stability of sulfonium-based AEMs (55, 65). Trostyanskaya and Makarova (55) investigated the chemical stability of ion exchange resins that contain ammonium, phosphonium and sulfonium cations. This study showed that when the resins were exposed to 4% aqueous NaOH for 24 h, the poly(benzyldialkylsulfonium chloride) ion exchange resin retained 40% of its water sorption capacity, while the poly(benzyltrialkylphosphonium chloride) ion exchange resin completely lost its water sorption capacity. This work suggests that chemical stability may follow the order: sulfonium > phosphonium. They also reported that the degradation of the benzyldialkylsulfonium cation yields dialkyl sulfide and benzylalkyl sulfide as byproducts. Prior work on the degradation of small molecule organic sulfonium cations shows that the main degradation reaction is the Stevens rearrangement reaction (42, 43) (Schemes 8 and 9), which is similar as the Stevens rearrangement of an ammonium cation (Scheme 3 and 4). To the authors’ best knowledge, the relatively poor chemical stability of the sulfonium cation in alkaline conditions, may be a primary reason that sulfonium-based AEMs have not been developed for the AFCs.

Scheme 8. Stevens rearrangement of a sulfonium cation (42)

Scheme 9. Stevens rearrangement of benzyldimethylsulfonium cation (42) 243 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Chemical Stability of Guanidinium-Based AEMs

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Recent studies (34, 35) show that guanidinium-based AEMs have good chemical stability in alkaline solutions. The conductivity was maintained after immersion in 1 M KOH solution at 60 °C for 48 h (34) or it reduced by only 30% after immersion in 0.5 M KOH at 80 °C solution for 382 h (35) (Table 2). Unfortunately, there are few investigations on the chemical stability of the small molecule organic guanidinium cation. A recent study showed that the degradation of guanidinium cation undergoes a SN2 reaction (Scheme 10) (66).

Scheme 10. Degradation of a guanidinium-based cation (66)

Chemical Stability of Imidazolium-Based AEMs There are far less studies on AEMs with cyclical cations (e.g., imidazolium, pyridinium, and quaternized DABCO, piperidinium) compared to AEMs with noncyclical cations (e.g., ammonium). Imidazolium-based AEMs are of interest due to the five-membered heterocyclic ring and π conjugated structure of the imidazolium cation. Several recent studies (30–32) developed imidazolium-based AEMs and reported good chemical stability based on conductivity results. For example, the conductivity of a 1-allyl-3-methylimidazolium-based random copolymer reduced by only 8% after immersion in 6 M NaOH solution at 60 °C for 120 h, and a 1vinyl-3-methylimidazolium-based block copolymer retained its conductivity after immersion in 1 M KOH solution at 60 °C for 400 h (See Table 2). However, a detailed stability analysis (e.g., degradation mechanisms) on these imidazoliumbased AEM were not conducted. However, a recent study (33) provides a thorough analysis of chemical stability for an imidazolium-based polymerized ionic liquid (PIL). A combination of conductivity and NMR experiments were utilized to comprehensively characterize and quantify the chemical stability of the AEM over a broad range of humidities, temperatures, and alkaline concentrations. In this study, high chemical stability was observed under dry conditions (10% RH) at 30 °C, humid and saturated conditions up to 80 °C, and even in mild alkaline conditions ([KOH] < 1 M) at 25 °C. Degradation was only observed under more vigorous conditions: dry conditions (10% RH) at 80 °C or at higher alkaline concentrations ([KOH] > 1 M). A ring-opening degradation pathway was suggested for the imidazolium cation based on a detailed analysis of the NMR spectra (Scheme 11). 244 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Scheme 11. Degradation of imidazolium-based AEMs (33)

The enhanced chemical stability of the imidazolium-based AEM was largely attributed to the steric hindrance and the presence of the π conjugated structure that reduce the SN2 and Hofmann elimination reactions. Furthermore, the deprotonation of a 1,3-alkyl-substituted imidazolium hydroxide can result in a relatively stable carbene. For example, stable imidazolium-based carbenes, such as 3-bis(adamantyl)imidazol-2-ylidene have been synthesized and isolated by others (67). It was reported that carbenes with large bulky substitutes are even more stable, and more importantly, the stable alkyl-substituted carbenes resulting from deprontonation can be reversely protonated by water (68). As a comparison, deprotonation of the tetraalkyl quaternary ammonium cation has been shown to result in the formation of a relatively unstable ylide intermediate and the ylide intermediate undergoes further degradation through direct degradation (Scheme 6) and/or ylide rearrangements, such as Stevens rearrangement (Scheme 4) and Sommelet-Hauser rearrangement (Scheme 5). The stable structure of imidazolium cations has also been demonstrated with a number of hydroxide-based ionic liquids (ILs) that have been synthesized and dried for use as catalysts (i.e., stable under dry conditions), such as 1-butyl3-methylimidazolium hydroxide (69) and 1-butyl-2,3-dimethylimidazolium hydroxide (70). As a comparison, ammonium counterparts, such as tetrabutylammonium hydroxide, are commercially available, but always stored in a liquid form at a low concentration (< 40 wt %) (i.e., requires solvation for stability) (71). In contrast to the work on ammonium and phosphonium cations, the chemical stability of small molecule imidazolium cations is relatively scarce. The degradation of purines, compounds that consist of a pyrimidine ring fused to an imidazole ring, appears to be the only relevant work reported. For example, several studies (72, 73) reported that 7-methylguanosine undergoes ring-opening reactions in a strong aqueous base due to a nucleophilic attack by the hydroxide anion (Scheme 12). One characteristic of the ring opening reaction is the formation of isomer products since there are two possible sites available in the imidazolium ring. This study showed the formation of isomers, which agrees with literature (74, 75). 245 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Scheme 12. Ring opening reaction of 7-methylguanosine under alkaline conditions (75)

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Chemical Stability of Pyridinium-Based AEMs There are a few studies on pyridinium-based AEMs (76–80). However, only several of these were for the application of the AFC (78, 79). Xiao et al. (78) synthesized a copolymer of vinylpyridinium and styrene, which revealed high conductivity (0.8 mS/cm at 25 °C), but poor fuel cell performance. They attributed this to a possible chemical degradation of the AEM. Several studies indicate that the benzyltrimethylammonium cation may have better chemical stability than the pyridinium cation (37, 79). The possible instability of the pyridinium cation may be attributed to the enhanced susceptibility for nucleophilic addition and displacement at the α- and γ-positions (Scheme 13) (81).

Scheme 13. Degradation reaction of a pyridinium cation under alkaline conditions followed by oxidization (81).

Chemical Stability of Other Cyclic Cation-Based AEMs The incorporation of other cyclic cations such as quaternized 1,4diazabicyclo(2,2,2)octane (DABCO), and piperidinium in AEMs has been also investigated in literature. Due to the chemical stability, the quaternized DABCO cation-based AEMs (37, 82) has received relatively more attention. It was reported that both Hofmann elimination reaction and nucleophilic displacement reaction can be reduced due to the merit of the DABCO molecule (37). The absence of Hofmann elimination was attributed to internal steric constraints from an anti-periplanar conformation of C(β)-H and C(α)-N. Moreover, the second nitrogen atom in the para position, reduces the acidity of the molecule and thus reduces the effect of the positively charged nitrogen facing the hydroxide group. The DABCO molecule process two amine groups and can be converted to either a mono-quaternized cation or a bis-quaternized ammonium cation. Bauer et al. indicated that the mono-quaternized one was much more stable (t1/2 = 42 min) than the bis-quaternized DABCO-based polymer (t1/2 = 2.3 min) in 2M KOH at 160 °C under nitrogen atmosphere. The higher degradation rate of bis-quaternized DABCO was attributed to a rapid elimination according to the 246 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

monomolecular E1 mechanism. For the degradation pathways, mono-quaternized DBCO undergoes a nucleophilic displacement reaction (Scheme 14), while the bis-quaternized DABCO cations convert into a piperazine structure (Scheme 15).

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Scheme 14. Degradation reaction of a mono-quaternized DABCO cation under alkaline conditions (37).

Scheme 15. Degradation reaction of a bis-quaternized DABCO cation under alkaline conditions (37).

Recent patented work (36) showed the IEC of the piperidinium-cation based AEMs can be well retained (> 90%) even at a high temperature (80 °C) in a strong alkaline condition (~ 1.8 M KOH). However, more investigation is needed to better understand the chemical stability of piperidium cation.

Chemical Stability of AEM Polymer Backbones The alkaline chemical stability of the polymer backbone in AEMs should also be considered. However, studies on AEM polymer backbone chemical stability have received less attention compared to the cation since it is generally assumed that it is more stable than the cation. Previous studies indicated that backbone instability can largely be related to an attack by the hydroxide anion (9, 83) or oxygen (38). For example, Slade et al. observed decreased CH2 signal and decreased nitrogen signal from 13C {1H} and 15N {1H} cross polarization and magic angle spinning (CP-MAS) NMR spectra, which was due to the degradation of the poly(vinylidene fluoride) (PVDF) backbone (9, 76). This instability in alkaline conditions was attributed to the attack of OH-, which resulted in an E2 elimination (Scheme 16). Other studies showed that the byproduct can undergo further degradation reactions, such as hydroxylation and carbonyl formation (84). In the presence of oxidants, the styrenic backbone may undergo oxidization and form carboxylic acid (Scheme 17) (15). 247 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

Scheme 16. Degradation of PVDF polymer backbone (1)

Scheme 17. Oxidization of polymer backbone (15)

Conclusions In summary, the chemical stability of AEMs poses a critical challenge that limits the wide scale use of AFCs. This instability stems from the highly nucleophilic and basic nature of the hydroxide anion. All covalently attached conducting cation groups that have been studied thus far are prone to some level of degradation by OH- attack, particularly at high alkaline concentrations and high temperatures. Generally, ammonium has shown better chemical stability compared to phosphonium and sulfonium, while heterocyclic cations, such as imidazolium, can enhance the chemical stability compared to non-cyclical cations due to their conjugated structure and steric hindrance. Recent studies on the development of AEMs for the AFC have primarily focused on measuring changes in ion exchange capacity and ionic conductivity under certain conditions over a period of time to assess chemical stability. Among these studies, the ammonium cation has received the most attention. Although more recent studies have shown enhanced stability (e.g., imidazolium-based AEMs), a deep fundamental understanding of AEM chemical stability is still lacking. In the future, the same approach as in the investigation of small molecule organic cations should be applied to AEMs (e.g., the use of advanced characterization techniques, such as nuclear magnetic resonance and isotope labeling). For example, understanding the electron-withdrawing nature of o-, and p-methoxy groups in phosphonium-based AEMs (e.g., benzyltri(2,4,6-trimethoxyphenyl) phosphonium cation) could aid in the design of new AEMs with high chemical stability. Overall, more fundamental investigations on the chemical stability of AEMs will be beneficial in the design of chemically robust AEMs that could result in long-lasting, high-performing AFCs.

248 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Acknowledgments The authors gratefully acknowledge the U.S. Army Research Office for the financial support (grant W911NF-07-1 0452, Ionic Liquids in Electro-Active Devices (ILEAD) MURI).

References

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

Couture, G.; Alaaeddine, A.; Boschet, F.; Ameduri, B. Prog. Polym. Sci. 2011, 36, 1521. Reid, J. H. U.S. Patent 736,016, 1903. Varga, Á. In Fuel cell electronics packaging; Kuang, K., Easler, K., Eds.; Springer: New York, 2007; p 5. Kordesch, K. V. In Fuel Cells; Mitchell, W., Jr., Ed.; Academic Press: New York, 1963; Vol. 1, pp 329−70. Rolla, A.; Sadkowski, A.; Wild, J.; Zoltowski, P. J. Power Sources 1980, 5, 189. Rikukawa, M.; Sanui, K. Prog. Polym. Sci. 2000, 25, 1463. Grot, W. G. U.S. Patents 3,718,627, 1968; 3,969,285, 1976; 4,026,783, 1977; 4,030,988, 1977. Agel, E.; Bouet, J.; Fauvarque, J. F. J. Power Sources 2001, 101, 267. Danks, T. N.; Slade, R. C. T.; Varcoe, J. R. J. Mater. Chem. 2003, 13, 712. Varcoe, J. R.; Slade, R. C. T. Fuel Cells 2005, 5, 187. Yan, J.; Hickner, M. A. Macromolecules 2010, 43, 2349. Jagur-Grodzinski, J. Polym. Adv. Technol. 2007, 18, 785. Yu, E. H.; Krewer, U.; Scott, K. Energies 2010, 3, 1499. Elabd, Y. A.; Hickner, M. A. Macromolecules 2011, 44, 1. Merle, G.; Wessling, M.; Nijmeijer, K. J. Membr. Sci. 2011, 377, 1. Zhou, J.; Ünlü, M.; Anestis-Richard, I.; Kohl, P. A. J. Membr. Sci. 2010, 350, 286. Robertson, N. J.; Kostalik, H. A., IV; Clark, T. J.; Mutolo, P. F.; Abruña, H. D.; Coates, G. W. J. Am. Chem. Soc. 2010, 132, 3400. Luo, Y.; Guo, J.; Wang, C.; Chu, D. J. Power Sources 2010, 195, 3765. Thomas, O. D.; Soo, K. J. W. Y.; Peckham, T. J.; Kulkarni, M. P.; Holdcroft, S. Polym. Chem. 2011, 2, 1641. Herman, H.; Slade, R. C. T.; Varcoe, J. R. J. Membr. Sci. 2003, 218, 147. Komkova, E. N.; Stamatialis, D. F.; Strathmann, H.; Wessling, M. J. Membr. Sci. 2004, 244, 25. Wu, Y.; Wu, C.; Xu, T.; Lin, X; Fu, Y. J. Membr. Sci. 2009, 338, 51. Wang, G.; Weng, Y.; Chu, D.; Xie, D.; Chen, R. J. Membr. Sci. 2009, 326, 4. Xu, H.; Fang, J.; Guo, M.; Lu, X.; Wei, X.; Tu, S. J. Membr. Sci. 2010, 354, 206. Pan, J.; Lu, S.; Li, Y.; Huang, A.; Zhuang, L.; Lu, J. Adv. Funct. Mater. 2010, 20, 312. Wang, J.; Wang, J.; Li, S.; Zhang, S. J. Membr. Sci. 2011, 368, 246. Zhang, Q.; Zhang, Q.; Wang, J; Zhang, S.; Li, S. Polymer 2010, 51, 5407. 249

In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

28. Guo, T.; Zeng, Q.; Zhao, C.; Liu, Q.; Zhu, A.; Broadwell, I. J. Membr. Sci. 2011, 371, 268. 29. Gu, S.; Cai, R.; Luo, T.; Chen, Z. W.; Sun, M. W.; Liu, Y.; He, G. H.; Yan, Y. S. Angew. Chem., Int. Ed. 2009, 48, 6499. 30. Guo, M.; Fang, J.; Xu, H.; Li, W.; Lu, X.; Lan, C.; Li, K. J. Membr. Sci. 2010, 362, 97. 31. Lin, B.; Qiu, L.; Lu, J.; Yan, F. Chem. Mater. 2010, 22, 6718. 32. Li, W.; Fang, J.; Lv, M.; Chen, C.; Chi, X.; Yang, Y.; Zhang, Y. J. Mater. Chem. 2011, 21, 11340. 33. Ye, Y.; Elabd, Y. A. Macromolecules 2011, 44, 5727. 34. Wang, J.; Li, S.; Zhang, S. Macromolecules 2010, 43, 3890. 35. Kim, D. S.; Kim, Y. S. ECS Trans. 2010, 33, 1867. 36. Daikoku, Y; Isomura, T.; Fukuta, K.; Yanagi, H.; Yamaguchi, M. Eur. Patent, EP 2,351,785 A1, 2011. 37. Bauer, B.; Tirathmann, H.; Effenberger, F. Desalination 1990, 79, 125. 38. Neagu, V.; Bunia, I.; Plesca, I. Polym. Degrad. Stab. 2000, 70, 463. 39. Zagorodni, A. A.; Kotova, D. L.; Selemenev, V. F. React. Funct. Polym. 2002, 53, 157. 40. Hofmann, A. W. Philos. Trans. R. Soc., London 1851, 141, 357. 41. Hofmann, A. W. Proc. R. Soc. London 1857, 9, 287. 42. Stevens, T. S.; Creighton, E. M.; Gordon, A. B.; MacNicol, M. J. Chem. Soc. 1928, 3193. 43. Stevens, T. S. J. Chem. Soc. 1930, 2107. 44. Johnstone, R. A. W.; Stevens, T. S. J. Chem. Soc. 1955, 4487. 45. Puterbaugh, W. H.; Hauser, C. R. J. Am. Chem. Soc. 1964, 86, 1105. 46. Hauser, C. R.; Kantor, S. W. J. Am. Chem. Soc. 1951, 73, 1437. 47. Schöllkopf, U.; Ludwig, U.; Ostermann, G.; Patsch, M. Tetrahedron Lett. 1969, 10, 3415. 48. Sommelet, M. Compt. Rend. 1937, 205, 56. 49. Kantor, S. W.; Hauser, C. R. J. Am. Chem. Soc. 1951, 73, 4122. 50. Hauser, C. R.; Van Eenam, D. N. J. Am. Chem. Soc. 1957, 79, 5512. 51. Pine, S. H.; Sanchez, B. L. Tetrahedron Lett. 1969, 10, 1319. 52. Jones, F.; Hauser, C. R. J. Org. Chem. 1961, 26, 2979. 53. A. R. Lepley, A. R.; Becke, R. H. Tetrahedron 1965, 21, 2365. 54. Chempath, S.; Boncella, J. M.; Pratt, L. R.; Henson, N.; Pivovar, B. S. J. Phys. Chem. C 2010, 114, 11977. 55. Trostyanskaya, E. B.; Makarova, S. B. Zhurnal Prikladnoi Khimii 1966, 39, 1754. 56. Gu, S.; Cai, R.; Luo, T.; Jensen, K.; Contreras, C.; Yan, Y. ChemSusChem 2010, 3, 555. 57. Arges, C. G.; Kulkarni, S.; Baranek, A.; Pan, K.-J.; Jung, M.; Patton, D.; Mauritz, K. A.; Ramani, V. ECS Trans. 2010, 33, 1903. 58. Hofmann, A. W.; Cahours, A. Philos. Trans. R. Soc., London 1857, 147, 575. 59. Fenton, G. W.; Ingold, C. K. J. Chem. Soc. 1929, 2342. 60. Hey, L.; Ingold, C. K. J. Chem. Soc. 1933, 531. 250 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by OHIO STATE UNIV LIBRARIES on May 31, 2012 | http://pubs.acs.org Publication Date (Web): May 7, 2012 | doi: 10.1021/bk-2012-1096.ch014

61. McEwen, W. E.; VanderWerf, C. A.; Zander, M. J. Am. Chem. Soc. 1959, 81, 3806. 62. Keldsen, G. L.; McEwen, W. E. J. Am. Chem. Soc. 1978, 100, 7312. 63. Wang, Z. Comprehensive Organic Name Reactions and Reagents, 1st ed.; Wiley-Interscience: 2009. 64. McEwen, W. E.; Axelrad, G.; Zanger, M.; VanderWerf, C. A. J. Am. Chem. Soc. 1965, 87, 3948. 65. Dewi, E. L.; Oyaizu, K.; Nishide, H.; Tsuchida, E. J. Power Sources 2003, 115, 149. 66. Kim, D. S.; Labouriau, A.; Guiver, M. D.; Kim, Y. S. Chem. Mater. 2011, 23, 3795. 67. Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc. 1991, 113, 361. 68. Amyes, T. L.; Diver, S. T.; Richard, J. P.; Rivas, F. M.; Toth, K. J. Am. Chem. Soc. 2004, 126, 4366–4374. 69. Ranu, B. C.; Banerjee, S. Org. Lett. 2005, 7, 3049. 70. Mehnert, C. P.; Dispenziere, N. C.; Schlosberg, R. H. U.S. Patent 6,552,232, 2003. 71. Kagimoto, J.; Fukumoto, K.; Ohno, H. Chem. Commun. 2006, 2254. 72. Hecht, S. M.; Adams, B. L.; Kozarich, J. W. J. Org. Chem. 1976, 41, 2303. 73. Chetsanga, C. J.; Bearie, B.; Makaroff, C. Chem.-Biol. Interact. 1982, 41, 235. 74. Smith, C. W.; Rasmussen, R. S.; Ballard, S. A. J. Am. Chem. Soc. 1949, 71, 1082. 75. Chetsanga, C. J.; Makaroff, C. Chem.-Biol. Interact. 1982, 41, 235. 76. Sata, T.; Yamane, Y.; Matsusaki, K. J. Polym. Sci., Part A: Polym. Chem. 1998, 36, 49. 77. Sata, T.; Tsujimoto, M.; Yamaguchi, T.; Matsusaki, K. J. Membr. Sci. 1996, 112, 161. 78. Huang, A.; Xia, C.; Xiao, C.; Zhuang, L. J. Appl. Polym. Sci. 2006, 100, 2248. 79. Hnát, J.; Paidar, M.; Schauer, J.; Žitka, J.; Bouzek, K. J. Appl. Electrochem. 2011, 41, 1043. 80. Kang, M.-S.; Choi, Y. J.; Moon, S. H. AIChE J. 2003, 49, 3213. 81. Joule, J. A.; Mills, K. Heterocyclic Chemistry. Chapter 5. Pyridines: Reactions and Synthesis, 4th ed.; 2000. 82. Altmeier, P. U.S. Patents 5,746,917, 1998; 5,936,004. 83. Danks, T. N.; Slade, R. C. T.; Varcoe, J. R. J. Mater. Chem. 2002, 12, 3371. 84. Ross, G. J.; Watts, J. F.; Hill, M. P.; Morrissey, P. Polymer 2001, 42, 403.

251 In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; Page, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.